Autonomic Programming Paradigm for High Performance Computing
Total Page：16
File Type：pdf, Size：1020Kb
Autonomic Programming Paradigm for High Performance Computing Item Type text; Electronic Dissertation Authors Jararweh, Yaser Publisher The University of Arizona. Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author. Download date 28/09/2021 09:03:59 Link to Item http://hdl.handle.net/10150/193527 AUTONOMIC PROGRAMMING PARADIGM FOR HIGH PERFORMANCE COMPUTING By Yaser Jararweh A Dissertation Submitted to the Faculty of the DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING In Partial Fulfillment of the Requirements For the Degree of DOCTOR OF PHILOSOPHY In the Graduate College THE UNIVERSITY OF ARIZONA 2010 2 THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Yaser Jararweh entitled “AUTONOMIC PROGRAMMING PARADIGM FOR HIGH PERFORMANCE COMPUTING” and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy _________________________________________________________Date: 08/09/2010 Salim Hariri _________________________________________________________Date: 08/09/2010 Ali Akoglu _________________________________________________________Date: 08/09/2010 Janet Wang Final approval and acceptance of this dissertation is contingent upon the candidate's submission of the final copies of the dissertation to the Graduate College. I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement. _________________________________________________________Date: 08/09/2010 Dissertation Director: Salim Hariri 3 STATEMENT BY AUTHOR This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under the rules of the library. Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgement of the source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the author. SIGNED: Yaser Jararweh 4 ACKNOWLEDGEMENTS First of all, I’m grateful to the GOD, most mighty, most merciful, who created me and gave me the ability to think and search for the truth. I would like to thank my Ph.D. advisor, Professor Salim Hariri, for his continued support and guidance during the course of my Ph.D. studies. Professor Hariri has allowed me the freedom to pursue my research interests, and provided me with encouragement and patience. I have benefited from him not only research wisdom, but also research ethics that I treasure most. In addition, I would like to thank Professor Ali Akoglu, Professor Janet Wang, Professor Ivan Djordjevic, and Professor Ferenc Szidarovszky for serving on my written and oral committees and offering valuable suggestions to improve this dissertation. My fellow lab mates in Professor Hariri’s group, I am indebted to my friend, instructor, critique, and mentor Professor Youssif Al-Nashif, for all what he did for me during the course of my Ph.D. studies. Last but not least, I am indebted to my parents, brothers and sisters, my wife, and my daughter “Ayah” for always being there for me. My achievements would not have been possible without their emotional support and encouragement. 5 To my Parents and Family 6 TABLE OF CONTENTS LIST OF FIGURES .......................................................................................................... 8 LIST OF TABLES .......................................................................................................... 10 ABSTRACT ..................................................................................................................... 11 CHAPTER 1 INTRODUCTION .............................................................................................. 13 1.1 Problem Statement .................................................................................... 13 1.2 Research Objectives .................................................................................. 15 1.3 Research Challenges .............................................................................. 16 1.4 Dissertation Organization ...................................................................... 19 2 BACKGROUND AND RELATED WORK ................................................... 21 2.1 Autonomic Programing Paradigm ........................................................ 21 2.1.1 Autonomic Programing Paradigm Research ................................. 21 2.1.2 Autonomic Programing Paradigm Concepts ................................ 24 2.2 Self-Configuration and Self-Optimizing Management Techniques ...... 28 2.2.1 Self-Configuration Management Techniques ............................... 28 2.2.2 Self-Optimizing Management Techniques ................................... 31 2.3 Power Management Techniques for Computing Systems ..................... 33 2.2.1 Power Management for CPU-Only Systems ................................. 35 2.2.2 Power Management for GPU Based Systems ............................... 36 2.4 Human Cardiac Modeling and Simulation ............................................ 39 2.5 Case Based Reasoning (CBR) ............................................................... 43 3 PHYSICS AWARE OPTIMIZATION FOR LARGE-SCALE SCIENTIFIC AND MEDICAL APPLICATIONS ......................................................................... 48 3.1 Introduction ............................................................................................ 48 3.2 Physics Aware Optimization (PAO) Research ...................................... 49 3.2.1 Physics Aware Programming (PAO) ............................................ 50 3.3 POA for Human Heart Simulation......................................................... 55 3.3.1 Human Ventricular Epicardia Myocyte Model .............................. 55 3.3.2 AP Phase Detection Technique ...................................................... 58 3.3.3 POA Implementation of the Ten-Tusscher’s Simulation............... 60 3.4 Experimental Results and Evaluation .................................................... 61 7 TABLE OF CONTENTS – Continued 4 PERSONAL SUPERCOMPUTING AND GPU CLUSTER SYSTEM ............ 63 4.1 Introduction ............................................................................................ 63 4.2 GPU and CUDA Programming ............................................................. 65 4.3 Personal Supercomputing and GPUs ..................................................... 67 4.3.1 Personal Supercomputing Benefits ............................................... 69 4.4 GPU Cluster ........................................................................................... 71 4.4.1 GPU Cluster Architecture ............................................................. 72 4.5 GPU Clusters Challenges and Problems ................................................ 74 4.6 Current Solutions to Overcome GPU Challenges .................................. 78 4.6.1 Problems with the Current Solutions ............................................ 81 5 AUTONOMIC POWER AND PERFORMANCE MANAGEMENT FOR GPU CLUSTER ................................................................................................................... 83 5.1 Introduction ............................................................................................ 83 5.2 Motivational Example ............................................................................ 84 5.2.1 Exploiting GPU Cluster Architecture ........................................... 86 5.3 Modeling the GPU Cluster Managed Components ............................... 87 5.3.1 Modeling the Device Manager...................................................... 92 5.3.2 Modeling the Server Manager ...................................................... 94 5.4 Experimental Evaluation ...................................................................... 106 6 CONCLUSION AND FUTURE RESEARCH DIRECTIONS ........................... 120 6.1 Summary .............................................................................................. 120 6.2 Contributions ....................................................................................... 122 6.3 Future Research Directions .................................................................. 123 REFERENCES .............................................................................................................. 125 8 LIST OF FIGURES Figure 1-1: Worldwide IT spending. Source: http:idc 2006 ............................................. 14 Figure 2-1: Autonomic computing system [99] ................................................................ 28 Figure 2-2: Different optimization classified by knowledge exploited [3]....................... 32 Figure 3-1: An example of diffusion problems................................................................. 51 Figure 3-2: The cardiac action potential phases ............................................................... 57 Figure 3-3: PAO-based Algorithm.................................................................................... 60 Figure 4-1: GPU architecture